© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 18807
    Prev
    Next
    INDEX
    Explanations

    racial slurs in the text

    oai_token-act-pair · gpt-3.5-turbo

    instances of racial slurs and related discriminatory language

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chance
    -0.78
     Supply
    -0.72
    apt
    -0.70
     Aviation
    -0.69
    blogspot
    -0.68
    work
    -0.66
     Satellite
    -0.66
     Synd
    -0.63
     Mystery
    -0.63
    road
    -0.63
    POSITIVE LOGITS
     slur
    3.82
     slurs
    2.87
     gul
    1.17
     pronoun
    1.16
     spat
    1.14
     nib
    1.05
     gob
    1.03
    olester
    0.98
     derogatory
    0.97
     tir
    0.97
    Activations Density 0.061%

    No Known Activations